What does observation include

The observation

is one of the central data collection methods in empirical social research and describes the visual and / or auditory observation of events such as social or interactive processes and situations as well as courses of action. Observations can be categorized in various ways: according to a high or low degree of structuring, participating or non-participating, open or covert.

The observation

Observations are an essential part of daily life. In contrast to the everyday, scientific observations are systematic and objective. They are a specific approach to social phenomena in which data about the objects of investigation are not based on their direct information (questionnaire, narrative interview), but are collected indirectly by the researcher or by persons instructed by him.

The observation generally relates to forms of behavior, action and interaction, i.e. to overt, manifest phenomena. Nevertheless, the perception and absorption capacity of the researchers and the observers deployed play an important role.

Scientific observation is therefore very generally the attentive, planned and target-oriented perception of processes, events, behaviors of living beings (humans and animals) depending on certain situations. The aim of the observation is to capture the subject of the respective interest as precisely as possible. It is a fundamental method of data and fact gathering. at the time of the occurrence. Such an observation describes or reconstructs social reality a research question using a clearly describable method.

In addition to questioning, experiment and content analysis, observation is one of the four main empirical research methods.

Classification of forms of observation

A distinction must first be made between observation

  • artificially established situations (social psychological experiments, such as arranged group discussions, the course of which is analyzed) and those
  • more natural Situations (analysis of group structures in youth gangs) that have not been specifically arranged for the research purpose.

Observation methods can be further differentiated according to the degree of standardizationto which the logging of the observations is subject.

In the above-mentioned experimental settings, but also in natural situations such as classrooms, courtrooms, etc., standardized, detailed and specialized ones often come up Coding schemes in which the observing person makes entries (type, frequency and length of the participation of individual persons in group events). It must therefore be determined before the investigation what is to be observed, i.e. what appears to be important with regard to the research question (Categorization).

In this context of standardized observation, a more detailed distinction is then made between survey methods in which the individual plot lines of actors are surveyed, and those for whom summary assessments their actions / behavior can be carried out using a rating process.

A further distinction is made as to whether only certain phenomena (behaviors, properties, etc.) should be recorded from the outset, or whether there is a claim to classify all behaviors according to a certain scheme. The most well-known and still occasionally used category system was developed by Bales (1951) under the name "Interaction Process Analysis" released.

The degree of Pre-structuring However, the observation can decrease to such an extent that initially the claim is made to observe everything or - because this claim is of course not realizable - to develop the important and notable dimensions of observation from the examined phenomenon. For this purpose, detailed written or oral protocols of the observations are first made, which are only structured according to specific dimensions in the course of the investigation.

Another distinction relates to the degree to which the researcher is involved in the field of study. One speaks of non-participant observationif the researchers themselves are not an active part of the observation field. Often - especially in the area of ​​open observation in natural settings - B. is only possible through more or less active participation in the field of study (Participating observation). A special moral-ethical (but also methodological) problem arises from the fact that it must be decided and accountable whether an observation is made open, i.e. with the knowledge and consent of the observed, or else covered, i.e. without their knowledge. The main fields of application for participant observation are ethnological or ethnographic analyzes of particular social groups or situations.

See also Example of a systematic nonparticipating observation

Sources of error

They can either be with the observer, with the instrument, or with the situation. A big problem is the fact that social behavior is abstract, what is observed does not have to correspond to "reality".

Methodological problems are considered to be important control the observation performance (do the researchers only see what they want to see?) and - in the case of participant observation - the so-called going native, i.e. an over-identification with the field of study, which can ultimately lead to the abandonment of scientific analysis.

The control problem can possibly be reduced by supplementing the observation with other methods of data collection. Above all, we recommend not only intensive (self) training for the observers, but also one Checking inter- and intra-rater reliability.

There are also the so-called systematic errorthat occur already in the preparatory phase of the observation. (e.g. awkward selection of the form of observation). In order to avoid such errors, one should follow the theoretically planned observation Pre-test be made. It represents a sample which often leads to the revision of the schema and further training of the observer.

Errors by the observer arise from lack of training, lack of objectivity and through too high demands to the observer.

Steps of systematic observation

Operational definition of the observation category

For example, if you want to observe aggressive behavior, you must first determine what exactly you mean by "aggressive". The observable behavior can also be described as "kicking someone", "pulling someone's hair" or "shouting loudly at someone". The operationalization serves to ensure that every other person comes to the same observation result.

Make an observation

If you want to know exactly how often someone is aggressive, you would have to keep an eye on them. Of course, this is not possible in practice. A period of time is therefore set in which the behavior of the particular person is observed. But it is very important that the observation always takes place under the same conditions. After it has been precisely determined in which period and in which situation one observes, an observation sheet is drawn up on which the observable behaviors are entered in their frequency. The observation sheet makes it very easy to analyze the behavior.


Bales, R.F. (1951). Interaction Process Analysis. Chicago: Chicago University Press.

Friedrichs, J. (1985). Methods of empirical social research. Opladen.

Grumber, K. (1974). Observation. Stuttgart: Teubner.

Koenig, R. (1972). Observation and Experiment in Social Research. Cologne.

Mees, U. & Selg, H. (Eds.) (1977). Behavioral observation and behavior modification. Stuttgart: Velcro.


Bortz, Jürgen & Döring, Nicola (2006): Research Methods and Evaluation. Heidelberg: Springer.
Kromrey, Helmut (2009). Empirical Social Research. Stuttgart: Lucius.
Stigler, Hubert (1996). Methodology. Lecture notes. University of Graz.
WWW: http://www-gewi.kfunigraz.ac.at/edu/studium/stoffen/meth.doc (98-01-03)
Stangl, Werner (1997). On the methodology of science in educational science. "Werner Stangl's worksheets".
WWW: http://paedpsych.jk.uni-linz.ac.at/INTERNET/ARBEITSBLAETTERORD/Arbeitsblaetter.html
ILMES - Internet lexicon of the methods of empirical social research.
WWW: http://www.lrz-muenchen.de/~wlm/ilm_b3.htm (01-11-12)
Keyword questioning. Eval-Wiki: http://www.eval-wiki.org/glossar/Befragung (11-08-21)

Contents overview of research methods in psychology and education

content :::: message :::: news :::: imprint :::: data protection :::: author :::: copyright :::: quote ::::